By Yu Wang and Longfei Zhang
During his victory speech on April 26, 2016, Donald Trump accused Hillary Clinton of playing the ‘woman card’, and said that she would be a failed candidate if she were a man. Clinton fired back during her victory speech in Philadelphia and said that “If fighting for women’s health care and paid family leave and equal pay is playing the ‘woman card’, then deal me in”. The ‘woman card’ subsequently became the meme of the week and its effects are much debated. According to CNN, New York Times, Washington Post and The Financial Times, this exchange between the two presidential nominees signaled a heated general election clash over gender.
Here, we present an image-driven method to measure the impact of the ‘woman card’ exchange between Hillary Clinton and Donald Trump. Building from a unique dataset of the two candidates’ Twitter followers, we first examine the transition dynamics of the two candidates’ Twitter followers one week before the woman card controversy and one week after. Then we train a convolutional neural network to classify the followers’ gender and study how women in particular are reacting to the ‘woman card’ exchange.
How many people are we looking at? A Lot!
|Hillary Clinton||Donald Trump|
|Woman Card Point||Before||After||Before||After|
We also carefully track the mobility of the followers.
1.Mobility of Hillary Clinton’s Unfollowers
|Bernie Sanders||Donald Trump||Ted Cruz|
2.Mobility of Donald Trump’s Unfollowers
|Hillary Clinton||Bernie Sanders||Ted Cruz|
Our analysis finds that the percentage of female followers has increased and the percentage of female unfollowers has decreased after the controversy.
This suggests that the ‘woman card’ comment has made women more likely to follow Hillary Clinton, less likely to unfollow her.
Interestingly, the ‘woman card’ controversy has not affected the gender composition of Trump followers.